We develop and test the Policy-Academic-Career Outcome Trajectory Model to examine how targeted and untargeted training programs affect the development of career trajectories and the achievement of scientific milestones among biomedical scientists. The design targets recipients of the National Research Service Award (F31) predoctoral fellowships. This approach permits recruitment of sufficient numbers to permit analysis by race, ethnicity and gender, and by type of NRSA award. The primary source of data is the scientific curriculum vitae, which is used to construct scientific career histories from undergraduate education through current scientific activity. The longitudinal data allow the specification of survival models to test the effect of training interventions on time to scientific milestones such as first publication, entry into professional science, and promotions. We hypothesize that the training interventions work as designed to allow members of under-represented groups to achieve field- appropriate milestones similar to members of majority groups. Furthermore, we hypothesize that the more interventions experienced over the course of career development, the greater the accumulative advantage to the minority scientist. Finally, we will test the hypothesis that women-and especially women of color- benefit from training interventions to a lesser degree than men. PUBLIC HEALTH RELEVANCE: Racial, ethnic and gender diversity in the scientific and clinical public health labor force is necessary for the conduct of research and delivery of services to the entire population. This research will improve understanding of science policies that will facilitate the attraction, participation, retention and promotion of members of under-represented groups in science and medicine.